Payment Cards Anomaly Detection

Analysis of payment cards to identify unusual transactions or suspicious activities to prevent potential fraud using Power BI

Table of contents

• Description
• Methodologies
• Results
• Files
• Technologies
• Setup
• Usage
• Status
• Inspiration
• Contact

Description

The objective of this project is to analyze the reasonableness of transactions of a fictional company in the US and identify unusual or suspicious transactions to prevent potential risks of fraud, helping accounts payable/receivables, expenditure and audit departments save 50% time in the audit process and costs. The dataset contains 26,864 rows and 8 columns, providing details of the transactions. The rows include the following details:

• T# - transaction ID
• ACC# - account ID
• DATE - date of transaction
• MERCHANT - merchant of transaction
• CATEGORY
• AMOUNT - amount of transaction
• TAX - tax applied in the transaction
• CITY - city of merchant
• ZIP - zipcode of merchant
• STATE - state of merchant

Methodologies

• Import data into Power BI
• Perform data cleaning and data visualization to identify outliers of transactions and assess their reasonableness
  • Identify most important categories
  • Identify transactions above a certain threshold
  • Identify small amounts and even amounts
  • Analyze transactions with specific cities/merchants
  • Determine tax rate per transaction
  • Determine businesses to negotiate a discount with
  • Explore categories
• Compile a report including findings and outliers and communicate results to Controller

Results

• In IL, most transactions were spent for groceries, food and restarants
• Peoria had the most and highest transactions
• May and August had the most transactions
• Plumb supply company was the merchant that had the most transactions

Files

• Pcard data.txt - text file containing raw data
• Pcard Anomaly Detection.pbix - Power BI file containing cleaned data and data visualization
• Power BI Anomaly Detection Report.docx - project report containing analyses and results
• Pcard Anomaly Detection Project & Report.pdf - pdf file containing all project details from the Power BI file and report from the docx file

Technologies

Project is created with:

• MS Excel
• Power BI
• Windows

Setup

• Download all data files if necessary

Usage

• Run Pcard Anomaly Detection.pbix to see data imported and data visualization
• Open Power BI Anomaly Detection Report.docx to see project report or Power BI Anomaly Detection Project & Report.pdf to see all project details and visualization and report combined

Status

Project is finished.

Inspiration

Project was completed for the final assignment for Augustana College's Intermediate Accounting III.

Contact

Created by:

• Janie Le
• Maxwell Xu
• Cody Goforth
• My Pham - @mypham14 - feel free to contact me on my LinkedIn!